Jacob Coxon quits Anthropic: labs racing to self-improving SI

On 9 Sep 2026 Jacob Coxon (@hilbertspaess) said he had resigned from Anthropic after about three years across OpenAI and Anthropic, arguing frontier labs are racing to self-improving superintelligence and “gambling with our lives.” POLITICO Europe and The Verge covered the exit; Anthropic alignment lead Evan Hubinger publicly agreed the existential risk is earnest — estimating higher than a 10% chance AI could kill all humans within a decade and noting no plan yet for SI-aligned control. HN item 49619227 and Hindustan Times also amplified the thread. The episode stacks another high-profile safety-exit signal on top of recent rogue-agent incidents.